Downregulation of Steroid Receptor Coactivator-2 Modulates Estrogen-Responsive Genes and Stimulates Proliferation of MCF-7 Breast Cancer Cells

Home , BCL11B

Ingvild S. Fenne1,2, Thomas Helland1,2, Marianne H. Fla˚geng1,2, Simon N. Dankel1,2, Gunnar Mellgren1,2, Jørn V. Sagen1,2* 1 Department of Clinical Science, University of Bergen, Bergen, Norway, 2 Hormone Laboratory, Haukeland University Hospital, Bergen, Norway

Abstract The p160/Steroid Receptor Coactivators SRC-1, SRC-2/GRIP1, and SRC-3/AIB1 are important regulators of Estrogen Receptor alpha (ERa) activity. However, whereas the functions of SRC-1 and SRC-3 in breast tumourigenesis have been extensively studied, little is known about the role of SRC-2. Previously, we reported that activation of the cAMP-dependent protein kinase, PKA, facilitates ubiquitination and proteasomal degradation of SRC-2 which in turn leads to inhibition of SRC-2- coactivation of ERa and changed expression of the ERa target gene, pS2. Here we have characterized the global program of transcription in SRC-2-depleted MCF-7 breast cancer cells using short-hairpin RNA technology, and in MCF-7 cells exposed to PKA activating agents. In order to identify genes that may be regulated through PKA-induced downregulation of SRC-2, overlapping transcriptional targets in response to the respective treatments were characterized. Interestingly, we observed decreased expression of several breast cancer tumour suppressor genes (e.g., TAGLN, EGR1, BCL11b, CAV1) in response to both SRC-2 knockdown and PKA activation, whereas the expression of a number of other genes implicated in cancer progression (e.g., RET, BCAS1, TFF3, CXCR4, ADM) was increased. In line with this, knockdown of SRC-2 also stimulated proliferation of MCF-7 cells. Together, these results suggest that SRC-2 may have an antiproliferative function in breast cancer cells.

Citation: Fenne IS, Helland T, Fla˚geng MH, Dankel SN, Mellgren G, et al. (2013) Downregulation of Steroid Receptor Coactivator-2 Modulates Estrogen-Responsive Genes and Stimulates Proliferation of MCF-7 Breast Cancer Cells. PLoS ONE 8(7): e70096. doi:10.1371/journal.pone.0070096 Editor: Karin Dahlman-Wright, Karolinska Institutet, Sweden Received May 7, 2013; Accepted June 14, 2013; Published July 30, 2013 Copyright: ß 2013 Fenne et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: Funding for this project was provided by the Norwegian Cancer Society (https://kreftforeningen.no), The Western Norway Regional Health Authority (http://www.helse-bergen.no/forskning/sam arbeidsorganet), KG Jebsen Center for Diabetes Research (http://www.uib.no/diabetes), Connie Gulborg Jansens legat (http://www.uib.no/mofa/forskning/fond-og-legater), and Meltzerfondet, University of Bergen (http://meltzerfondet.no). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. * E-mail: [email protected]

Introduction progression. In contrast to SRC-2, SRC-1 and SRC-3 are frequently overexpressed in a subset of breast cancers, and The Steroid Receptor Coactivator (SRC) family, also known as correlate with a more aggressive tumour phenotype and poor p160 proteins, consist of the three members, SRC-1 (NCOA1) [1], prognosis and resistance to endocrine treatment [4,23–28]. Even SRC-2 (NCOA2/GRIP1/TIF2) [2,3], and SRC-3 (NCOA3/ though depletion of SRC-2 in MCF-7 cells has been shown to AIB1/ACTR/RAC-3/pCIP/TRAM-1) [4–8]. Even though decrease estrogen-dependent ERa transactivation function, loss of SRC-2 is functionally and structurally related to the other two SRC-2 does not seem to affect estrogen-dependent proliferation of SRC members, knockout studies in mice have shown that SRC-2 MCF-7 cells [29]. This is in contrast to studies of SRC-3-depleted plays distinct functional roles in fertility and ductal branching in MCF-7 cells in which the estrogen-dependent proliferation of the mammary gland [9,10], glucose- and lipid metabolism [11,12], cells was significantly reduced [29–32]. Molecular studies of each regulation of bone mass [13], cardiac function [14] and SRC in MCF-7 breast cancer cell suggest that the SRCs exhibit progesterone-dependent cell cycle and immunity [15]. The differential regulation of endogenous ER-target genes, indicating SRC-specific functions are believed to be due to their tissue- specific contributions of each SRC member to promote breast specific expression levels, different affinities for various NRs, cancer [17,29,33,34]. In addition, distinct PTMs of the SRC competition between NRs to recruit SRCs and between SRCs members play a crucial role in controlling their intracellular levels themselves for binding to NRs, and different post-translational and functions, which may have significant impact on breast modifications (PTMs) that regulate their protein levels and activity carcinogenesis and response to endocrine treatment in breast [16–18]. There is also evidence that SRC-2, in contrast to SRC-1 cancer patients [18]. In previous studies we have demonstrated and SRC-3, has repressive effects on specific ER- and glucocor- that activation of the cAMP-dependent protein kinase (PKA) ticoid receptor (GR) target genes in the presence of their respective signalling pathway targets SRC-2 coactivator function through its ligands [19–22]. While the roles of SRC-1 and SRC-3 have been ubiquitination and proteasomal degradation. This in turn inhibits extensively studied in breast cancer, less is known about the function of SRC-2 in regulating genes involved in breast cancer

PLOS ONE | www.plosone.org 1 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

SRC-2-mediated coactivation of ERa and modulates transcription of the ERa target gene pS2 [35–37]. In the present study, our aim was to characterize the role of SRC-2 on global expression of genes in MCF-7 breast cancer cells. We also wanted to explore the effects of PKA-induced degradation of SRC-2 on the expression of genes involved in breast tumourigenesis. Interestingly, our data suggest that SRC-2 is important for the expression of various ER-target genes linked to breast cancer progression, including specific oncogenes and tumour suppressor genes. A subset of these genes was also found to be modulated through PKA-induced downregulation of SRC-2. Moreover, proliferation data suggested that knockdown (KD) of SRC-2 stimulated proliferation of MCF-7 cells. Taken together, our results suggest that SRC-2 play an important role in regulating expression of a subset of ER-target genes involved in proliferation of MCF-7 cells.

Materials and Methods Cell Cultures MCF-7 human breast adenocarcinoma cells and Human embryonic kidney 293T (HEK 293T) cells (ATCC) were grown at 37 uC and 5% CO2, in Dulbecco’s modified Eagle’s medium (DMEM) (Cambrex, Verviers, Belgium) supplemented with 4,5 g/ liter glucose, 2 mM L-glutamine, 10% foetal bovine serum (FBS), 100 units penicillin, 100 mg streptomycin. The MCF-7 cell medium also contained 1 mM insulin. For experiments MCF-7 cells were seeded in phenol red-free DMEM supplemented with 5% charcoal-stripped FBS.

Short Hairpin RNA (shRNA) lentiviral Transduction Five individual lentiviral pLKO.1-puro short hairpin (sh) RNA plasmids targeting different sequences on SRC-2/GRIP1 or SRC- 3/AIB1 mRNA and a pLKO.1-puro empty vector control were purchased from Sigma MissionH RNAi (Sigma). To produce Figure 1. Downregulation of SRC-2 in MCF-7 cells induces lentiviral stocks each plasmid was cotransfected with the lentiviral distinct changes in global gene expression profiles. (A). packaging plasmids psPAX2 and pMD2G into individual plates Quantification of SRC-2 mRNA expression in shRNA lentivirus-infected with HEK 293T cells using SuperFect (QIAGEN, Valencia, CA). MCF-7 cells. mRNA levels of SRC-2 in a MCF-7 cell line infected with shRNA targeting SRC-2 (SRC-2 shRNA) were compared to the expression The virus containing cell culture supernatants from each culture in a control shRNA MCF-7 cell line (Ctr shRNA), and in a MCF-7 cell line were collected 48 h after transfection and used for lentiviral transduced with shRNA lentivirus targeting SRC-3 (SRC-3 shRNA). The transduction of individual MCF-7 cells cultures in the presence of mRNA expression of SRC-2 is relative to TBP mRNA. The results are 10 mg/ml mg polybrene per ml virus supernatant. Two days after representative of at least three independent experiments. (B). Western infection 1 mg/ml puromycin was added in order to select for blotting analyses of SRC-2-depleted MCF-7 cells. MCF-7 cells infected infected cells. The puromycin selections were maintained for three with shRNA lentivirus targeting SRC-2 (SRC-2 shRNA) or a negative control shRNA empty vector (Ctr shRNA), were grown in phenol red-free weeks to obtain cells containing stably integrated shRNA. DMEM supplemented with charcoaled stripped FBS (5%) and 17b- estradiol (10 nM) for two days. The Ctr shRNA cells were then treated RNA Extraction and Quantitative Real-time PCR with either Vehicle or 8-CPT-cAMP (150 mM), IBMX (50 mM) and forskolin Total RNA from MCF-7 cells was extracted using RNeasy Mini (10 mM) (cAMP) for 24 hours. Immunoblotting was performed with anti- kit (Qiagen, CA). 1 mg RNA was reverse transcribed using the TIF2 antibody and anti-GAPDH antibody. The results shown are representative of at least three independent experiments. (C). Micro- cDNA synthesis kit (Roche Basel, Switzerland). The real time array analyses of five RNA samples isolated from five individual cell quantitative reverse transcription (qRT)-PCR analyses were samples of shRNA control MCF-7 cells (Ctr shRNA), SRC-2 KD MCF-7 cells carried out using the LightCyclerH RNA Master SYBR Green I (SRC-2 shRNA) and control shRNA cells treated with cAMP elevating kit in a LightCycler rapid thermal cycler system (Roche, Basel, agents, as described in A. A Venn diagram shows the number of Switzerland). The mRNA expression levels of target genes were individual and overlapping sets of genes differentially expressed after quantified relative to the housekeeping gene TATA-binding SRC-2 KD (SRC-2 shRNA) and after treatment with cAMP elevating agents. To examine which genes were similarly differentially expressed protein (TBP). The primer sequences are provided in the Materials between the two treated groups when compared to control, a SAM and Methods Supporting Information (Table S1). analysis with overlapping genes was performed. The fold change cut-off value $1.5, and q-value = 0, was used to determine differentially Microarray Preparation of Samples expressed genes. The shRNA-expressing MCF-7 cells were grown for three days doi:10.1371/journal.pone.0070096.g001 in phenol red-free DMEM with 5% charcoal-stripped FBS and 10 nM 17b-estradiol. RNA was extracted from five independent plates of cells (replicates) expressing the pLKO.1-puro empty shRNA vector (Control shRNA), from five replicates of SRC-2

PLOS ONE | www.plosone.org 2 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

Table 1. Overlapping decreased expression changes relative to control induced by SRC-2 KD and cAMP in MCF-7 cells.

Signal intensity FC

SRC-2 Ctr SRC-2 shRNA/ cAMP/Ctr Gene Definition shRNA cAMP shRNA Ctr shRNA shRNA

AKR1B10 aldo-keto reductase family 1, member B10 (aldose reductase) 255 257 507 21.98 21.98 ARHGEF19 Rho guanine nucleotide exchange factor (GEF) 19 420 415 632 21.50 21.53 BCL11B B-cell CLL/lymphoma 11B (zinc finger protein), 232 313 570 22.46 21.83 transcript variant 1 CAV1 caveolin 1, caveolae protein, 22kDa 3534 2946 5595 21.59 21.91 EGR1 early growth response 1 1159 1125 1799 21.55 21.62 KLK5 kallikrein-related peptidase 5, transcript variant 1 345 292 598 21.74 22.06 KLK5 kallikrein-related peptidase 5, transcript variant 2 244 226 416 21.72 21.84 KRT1 keratin 1 (epidermolytic hyperkeratosis) 143 118 229 21.60 21.94 KRT5 keratin 5 171 197 347 22.03 21.76 KRTDAP keratinocyte differentiation-associated protein 173 283 571 23.29 22.02 Protein S100-A9 S100 calcium binding protein A9 (calgranulin B) 308 644 1345 24.37 22.11 Protein S100-A8 S100 calcium binding protein A8 242 530 1892 27.82 23.58 TAGLN transgelin, transcript variant 2 257 350 797 23.10 22.28

Genes with FC $1.5 and q-value = 0 are shown in addition to their respective expression signals in the three microarray samples. cAMP, short-hairpin control MCF-7 cells treated with cAMP analogue (8-CPT-cAMP) and cAMP-elevating agents (IBMX and forskolin). doi:10.1371/journal.pone.0070096.t001 shRNA expressing cells (SRC-2 shRNA), and from five replicates J-Express were performed on quantile normalized and logarith- of Control shRNA cells treated with 150 mM 8-parachlorophe- mically transformed (base 2) signal intensity values [38]. Corre- nylthio-cAMP (8-CPT-cAMP), 50 mM 3-isobutyl-1-methylxan- spondence Analysis (CA) and hierarchical clustering with Pearson thine (IBMX), and 10 mM forskolin for 24 hours. Sample Correlation as a distance measure were applied to study global preparations were balanced and randomized in each step of lysate trends in the data and search for outliers within the sample groups collection, RNA extraction and labelling, and microarray hybrid- [39]. Differentially expressed genes were detected through the ization. Microarray was performed using the Illumina HumanRef- Significance Analysis of Microarray (SAM) [40], and defined by 8 v 3.0 Expression BeadChips. fold change of at least 1.3 and q-value = 0. To search for over- represented functional categories among the differentially ex- Microarray Analysis Using the Illumina Iscan System pressed genes, Protein ANalysis THrough Evolutionary Relation- 400 ng of total RNA from each cell sample (three biological ships (PANTHER) (version 7, http://www.pantherdb.org) was groups, five samples within each group, 15 samples total) was used to organize differentially expressed genes in categories biotin-labelled and amplified using the Illumina TotalPrep RNA representing biological functions and molecular functions. The amplification kit (version 0606, AmbionH, USA) and the Bonferroni correction for multiple testing was used to calculate p- Eppendorf Mastercycler (EppendorfH, Germany). Quality and values for the over-represented categories. quantity measurement of the biotin-labelled cRNA were performed using the Agilent 2100 Bioanalyzer and the NanoDropH Western Blotting ND-1000. 750 ng of cRNA was thereafter hybridized to the Procedures for Western blotting are previously described HumanRef-8 v.3.0 Expression Bead Chips containing gene- (Hoang et al 2004). Primary antibodies used in the immunoblot- specific probes at 58uC for 17 hours. The HumanRef-8 v.3.0 ting experiments were mouse monoclonal anti-TIF2 (BD Biosci- Expression BeadChip targets approximately 24500 genes derived ences, San Jose, CA) and mouse monoclonal anti-GAPDH from human genes in NCBI RefSeq database. The hybridization (Chemicon International, Temecula, CA). was performed according to the Whole – Genome Gene Expression Direct Hybridization Assay Guide (Illumina Inc.). Cell Proliferation Assay The fluorescence of the biotin-labelled cRNA was measured using Cell proliferation assays were performed using the xCELLi- the iScan Reader (Illumina Inc.). Analyses were performed at the gence system Real-Time Cell Analyzer (RTCA) DP instrument, as Norwegian Microarray Consortium (NMC) Core Facility, Uni- described by the manufacturer (Roche, Basel, Switzerland). MCF- versity of Bergen, Norway. 7 cells were seeded in phenol red-free DMEM supplemented with 5% charcoal-stripped FBS for three days prior to the experiments. Microarray Data Extraction and Analysis Background impedance was determined by incubating E-Plates The raw data from the microarray was imported into with 100 mL cell medium for 30 minutes at 37uC and 5% CO2. GenomeStudio Data Analysis Software (Illumina, Inc.) for quality 86103 control shRNA cells and SRC-2 shRNA cells were then controls. The control probes were then removed and a text file seeded into individual wells of the E-plates and treated with 10 nM containing the signal and detection p-values per probe for all 17b-estradiol or Vehicle. To activate the PKA signalling pathway, samples was created and imported to J-Express Pro software the cells were treated with 8-CPT-cAMP (150 mM), IBMX version 2009 (MolMine, Norway). Quality controls and analyses in (50 mM) and forskolin (10 mM). The plates containing the cells

PLOS ONE | www.plosone.org 3 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

when analysing the whole human NCBI genome (19,911 genes), e.g. 31% of the 19,911 genes belong to the ‘‘cellular process’’ category. The colour intensity scales are based on the statistical significance (-log p-value) of over- and under-represented PANTHER functional categories. Red colour displays over-represented categories and blue colour represents under- represented categories. A specific category will display red colour (overrepresentation) when there is a higher percentage of genes in the dataset (SRC-2 shRNA and cAMP vs. Ctr shRNA) compared to the percentage of genes in the reference column mapping to this specific category. E.g. 53% of the 70 downregulated genes map to the category ‘‘cellular process’’ while 31% of the reference genes map to this category, hence this category is over-represented among the downregulated genes and displays a red colour. The opposite principle will be true for the under- represented categories (blue colour). Ref, reference. Arrow up, upregulated genes. Arrow down, downregulated genes. doi:10.1371/journal.pone.0070096.g002

were then incubated at room temperature for 30 minutes before placing them into the RTCA-unit. Cell growth was measured for 120 hours by monitoring the impedance every 15 minutes. Impedance is represented by cell index (CI) and was calculated as follows: CIJ(Zi_Z0)/15W, where Zi is the impedance at an individual time-point, and Z0 is the background impedance. Average CI was calculated from a minimum of 2–4 wells per time- point and per experiment.

Results Loss of SRC-2 in MCF-7 Breast Cancer Cells Induces Distinct Changes in the Global Gene Expression Previously we have shown that activation of PKA leads to increased ubiquitination and subsequent degradation of SRC-2, which in turn leads to inhibition of ERa transactivation function [35,36]. Here we wanted to study the role of SRC-2 and the functional relevance of the PKA-mediated regulation of SRC-2 on gene expression in MCF-7 human breast cancer cells. Thus, we analysed global gene expression profiles of MCF-7 cells after either KD of SRC-2 or treatment with cAMP elevating agents. As shown in Figure 1A, there was approximately 65% reduction in SRC- 2 mRNA expression after KD of SRC-2 (SRC-2 shRNA) compared to the control shRNA MCF-7 cells (Ctr shRNA), which was not found in a cell line expressing shRNA against SRC-3. Western blotting analyses confirmed successful KD of SRC-2 (Figure 1B). In order to activate PKA in the Ctr shRNA cells, they were treated with a cAMP analogue (8-CTP-cAMP) and cAMP elevating agents (IBMX and forskolin) for 24 hours. This time point was optimal to achieve degradation of SRC-2. As previously shown [36,41], activation of PKA resulted in a significant reduction in SRC-2 protein level. Five samples of RNA from each of the three different treatments (Ctr shRNA, SRC-2 shRNA and cAMP) were subjected to microarray analysis. Correspondence Analysis which displays the differences in global gene expression in a two-dimensional plot revealed a distinct separation between the three groups and no outliers, indicating differential expression of genes at the global level (Figure S1). Using a fold change $1.5 and a q-value = 0, a total number of 383 genes (194 downregulated and 189 Figure 2. Functional categorization of differentially expressed upregulated) were differentially expressed in the SRC-2 KD cells overlapping genes. PANTHER was used to search for over-represented (Tables S2 and S3), whereas the cAMP-treated cells showed 210 Biological Process categories (A) and Molecular Function categories (B) among the differentially expressed overlapping up-regulated genes after differentially expressed genes (68 downregulated and 142 SRC-2 KD (SRC-2 shRNA) and after exposure of cAMP elevating agents, and upregulated) (Figure 1C), (Tables S4 and S5). among the differentially expressed overlapping downregulated genes from the same two samples (fold change $1.3, q-value = 0). Bonferroni Overlapping Gene Expression Changes Induced by SRC-2 correction for multiple testing was performed and a p-value of 0.05 was chosen as inclusion criterion for functional categories. The numbers in the Knockdown and cAMP figure are percentage numbers. I.e. the reference column at the left of the In order to study genes that could potentially be regulated table displays the percentage of genes that belongs to a specific category through PKA-mediated downregulation of SRC-2, we searched

PLOS ONE | www.plosone.org 4 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

Table 2. Overlapping increased expression changes relative to control induced by SRC-2 KD and cAMP in MCF-7 cells.

Signal Intensity FC

Gene Definition SRC-2 cAMP Ctr shRNA SRC-2 shRNA/ cAMP/Ctr shRNA Ctr shRNA shRNA

ADM adrenomedullin 336 270 176 1.90 1.54 AKR1C2 aldo-keto reductase family 1, member C2), transcript variant 1 10519 10085 6693 1.57 1.51 BCAS1 breast carcinoma amplified sequence 1 474 548 229 2.07 2.39 CXCR4 chemokine (C-X-C motif) receptor 4, transcript variant 2 411 285 187 2.19 1.51 C9orf152 chromosome 9 open reading frame 152 776 904 446 1.74 2.03 FAM46A family with sequence similarity 46, member A 1647 3923 976 4.00 4.00 LYPD6B (hypothetical protein LOC130576), LY6/PLAUR 1361 1205 733 1.86 1.64 domain containing 6B NUCB2 nucleobindin 2 625 778 413 1.50 1.87 RET ret proto-oncogene, transcript variant 4 986 889 538 1.83 1.65 RET ret proto-oncogene, transcript variant 2 1045 1077 524 1.99 2.04 S100P S100 calcium binding protein P 3980 5622 1275 3.12 4.41 TFF3 trefoil factor 3 (intestinal) 17129 17494 10193 1.68 1.72 TFPI tissue factor pathway inhibitor, transcript variant 2 537 606 355 1.10 1.19 TFPI tissue factor pathway inhibitor, transcript variant 1 416 482 289 1.51 1.70

Genes with FC $1.5 and q-value = 0 are shown in addition to their respective expression signals in the three microarray samples. cAMP, short-hairpin control MCF-7 cells treated with cAMP analogue (8-CPT-cAMP) and cAMP-elevating agents (IBMX and forskolin). doi:10.1371/journal.pone.0070096.t002 for overlapping differentially expressed genes after SRC-2 KD or as cellular component organization, whereas the expression of elevation of cAMP. As shown in the Venn diagram in Figure 1C, genes involved in cell communication was enhanced. the expression of 12 genes decreased, whereas 12 genes showed increased expression both after SRC-2 KD or cAMP treatment Depletion of SRC-2 Downregulates Tumour Suppressor (Table 1 and Table 2). Genes and Upregulates Oncogenes To gain insight into the functions of the genes changed by both To confirm that gene expression changes observed in the SRC-2 depletion and cAMP elevation, we performed gene microarray data were representative of the original samples, a ontology analyses of the overlapping genes using the PANTHER subset of the differentially expressed overlapping genes, were classification system. We observed that several of these genes were selected for validation by qRT-PCR analysis (Table 3). Based on involved in the PANTHER Biological Process categories Cell cycle the fact that little is known about the role of SRC-2 in breast (e.g. RET, S100P, KLK5, S100A8, S100A9, and EGR1), Cell motion tumourigenesis, genes that have been associated with breast cancer CXCR4 RET TFF3 S100P S100A8 S100A9 (e.g. , , , and ), Immune were selected for validation by qRT-PCR. The five genes showing response (e.g. CXCR4, S100P, S100A8, S100A9, TFF3, KLK5, RET decreased expression after SRC-2 KD and PKA activation and TFP1), Signal transduction (e.g. CXCR4, RET, TFF3, S100P, compared to control, EGR1, TAGLN, CAV1, BCL11b and S100A8, S100A9 and CAV1) and Metabolic processes (e.g. RET, AKR1B10, together with five overlapping genes showing increased TFP1, AKR1C2, KLK5, AKR1B10, BCL11B and EGR1) (data not expression, RET (tv2 and tv4), BCAS1, TFF3, ADM, CXCR4, were shown). In order to increase the number of overlapping genes to be chosen for validation. EGR1, CAV1, TAGLN and BCL11b are included in the PANTHER analyses but simultaneously maintain estrogen-responsive genes described as tumour suppressors in a low chance of false positive hits, fold change cut-off value was breast cancer [42–45], whereas the upregulated genes are lowered from 1.5 to 1.3 (q-value = 0). Our analyses revealed 124 estrogen-responsive genes described as breast cancer oncogenes overlapping genes by which 75 were decreased and 45 genes (RET, BCAS1, TFF3) [46–48] or known to stimulate breast cancer showed increased expression after SRC-2 KD and activation of metastasis and progression (CXCR4, ADM) [49,50]. We observed PKA (Tables S6 and S7). As shown in Figure 2A, PANTHER that the mRNA expression of these genes was reversed by functional analyses of the downregulated overlapping genes revealed a significant overrepresentation of the Biological Process overexpression of SRC-2 in MCF-7 cells (data not shown). As categories referred to as Cellular process and Cellular component shown in Table 3, the qRT-PCR validation results were highly organization compared to control MCF-7 cells (Ctr shRNA). consistent with the microarray results, suggesting that SRC-2 is Analysing the overlapping upregulated genes we observed a clear regulating estrogen-responsive genes involved in breast tumour- overrepresentation of the Biological Process category Cell com- igenesis. The results indicate that downregulation of SRC-2 munication (Figure 2A). PANTHER analyses of Molecular mediates decreased expression of several tumour suppressor genes, Function categories showed that genes belonging to Structural whereas the expression of genes involved in oncogenesis were molecular activity were overrepresented amongst the downregu- increased. lated genes by SRC-2 KD- and cAMP (Figure 2B). This suggest that PKA stimulation and SRC-2 silencing in MCF-7 cells entails decreased expression of genes involved in cellular process as well

PLOS ONE | www.plosone.org 5 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

Figure 3. PKA-mediated downregulation of SRC-2 changes mRNA expression of ER-target genes. Control MCF-7 cells (Ctr shRNA) and SRC-2 KD MCF-7 cells (SRC-2 shRNA) were grown in phenol red-free DMEM supplemented with charcoaled stripped FBS (5%) and 17b-estradiol (10 nM) for two days. The two cell lines were then treated with Vehicle or 8-CPT-cAMP (150 mM), IBMX (50 mM) and forskolin (10 mM) (cAMP) for 24 hours. The mRNA expression of selected genes was measured by qRT-PCR. The expression level of each gene is relative to TBP mRNA. The results presents mean values 6 SE obtained from four-six independent qRT-PCRs. doi:10.1371/journal.pone.0070096.g003

Estrogen-responsive Genes Regulated Through cAMP/ via an SRC-2 independent pathway (Figure 3). The relative PKA- PKA-mediated Degradation of SRC-2 induced downregulation of EGR1, BCL11b and TAGLN, but not Next we wanted to verify whether the expression of the selected CAV-1 was counteracted by SRC-2 KD. Together, these results overlapping genes was regulated through PKA-mediated degra- suggested that expression of BCAS1, RET, EGR1, BCL11b and dation of SRC-2. Thus, qRT-PCR was used to quantify the TAGLN are regulated through PKA-induced SRC-2 degradation mRNA expression changes of selected genes in control MCF-7 (Figure 3), whereas PKA regulates the expression of TFF3 and cells and in SRC-2-depleted cells in response to treatment with CAV1 independently of SRC-2 degradation. PKA activating agents (Figure 3). The relative increase in mRNA expression of BCAS1, RET (tv2) and RET (tv4) due to PKA Depletion of SRC-2 Stimulates Breast Cancer Cell activation (cAMP) observed in the control shRNA cells was Proliferation diminished or absent in the SRC-2 shRNA cells. In cells with Since our results indicated that KD of SRC-2 changes the reduced SRC-2 level, adding PKA activating agents did not result expression of estrogen-responsive genes known to be involved in in any further increase in the expression of these three genes, carcinogenesis, we wanted to examine whether KD of SRC-2 suggesting that the cAMP effect is mediated via downregulation of affected the real time growth of MCF-7 cells by using the SRC-2. In contrast, TFF3 mRNA levels were increased by PKA in xCELLigence System. We also examined the growth of control both cell lines, suggesting that this gene is also regulated by PKA shRNA cells and SRC-2 shRNA cells treated with cAMP analogue

PLOS ONE | www.plosone.org 6 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

Table 3. Validation of selected genes by qRT-PCR.

Microarray Q-rt-PCR

SRC-2 shRNA/Ctr SRC-2 shRNA/Ctr Gene shRNA cAMP/Ctr shRNA shRNA 95% CI cAMP/Ctr shRNA 95% CI

ADM 1.90 1.54 2.81 1.16–6.81 2.10 0.95–4.63 BCAS1 2.07 2.39 4.21 1.74–10.20 5.11 2.74–9.54 CXCR4 2.19 1.51 4.97 2.40–10.30 2.55 1.89–3.44 RET (tv4) 1.839 1.65 2.53 1.22–5.18 2.14 1.17–3.89 RET (tv2) 1.20 2.04 1.88 0.75–4.38 1.63 0.88–3.02 TFF3 1.68 1.72 2.46 1.06–5.72 2.44 1.46–3.78 AKR1B10 21.98 21.98 0.46 0.24–0.88 0.39 0.27–0.55 BCL11B 22.45 21.83 0.20 0.11–0.40 0.44 0.24–0.80 CAV1 21.59 21.91 0.57 0.33–0.99 0.45 0.37–0.56 EGR1 21.55 21.62 0.82 0.41–1.66 0.66 0.34–1.27 TAGLN 23.10 22.28 0.22 0.14–0.34 0.35 0.28–0.44

Fold change values from the microarray analyses compared to quantification of mRNA expression by qRT-PCR presented as geometric mean with 95% confidence intervals (CI), n = 5. cAMP, short-hairpin control MCF-7 cells treated with cAMP analogue (8-CPT-cAMP) and cAMP-elevating agents (IBMX and forskolin). doi:10.1371/journal.pone.0070096.t003 and cAMP-elevating agents. The cell proliferation was monitored associated with improved prognosis [54]. SRC-2 is implicated in both in the absence and presence of 17b-estradiol. Interestingly, various cancers including colon, prostate, endometrial, liver, and MCF-7 cells with reduced level of SRC-2 showed a significant astrocytic brain cancer [53,55–58]. In breast tumour tissue, increase in cell proliferation compared to the control shRNA cell endocrine therapy has also been shown to induce the expression line. This was observed both in the presence and absence of 17b- SRCs [59,60]. Still, there are few reports regarding the estradiol (Figures 4A and 4B). Moreover, we observed that MCF-7 contribution of SRC-2 in cell growth and its role in regulating cell growth increased significantly after treatment with the PKA- genes involved in cell proliferation and cancer progression. Our activating agents. The cAMP-stimulated growth was also observed findings suggest an inhibitory role of SRC-2 in breast tumourigen- in the SRC-2 KD cells (Figures 4A and 4B). MCF-7 cells treated esis which differs from the established oncogenic function of two with both SRC-2 shRNA and PKA-activating agents showed the other SRC family members. A recent study demonstrated that most pronounced cell proliferation, suggesting that PKA has an SRC-3, but not SRC-2, is required for estradiol-dependent growth effect on proliferation independent of SRC-2 degradation. of breast cancer cells, which is in agreement with our observations Together, these data suggest that downregulation of SRC-2 [29]. Another report have shown that SRC-3, in contrast to SRC- induce proliferation of MCF-7 cells. 2, stimulates proliferation of androgen-dependent and androgen- independent prostate cancer cell and tumour growth [61], Discussion indicating a different role of SRC-2 in these types of cancer. The reason for these apparent differences in growth regulation Several studies have examined how the different members of the between SRC family members is not clear. However, it has been SRC coactivator family promote carcinogenesis. The three SRCs shown that the different SRCs have tissue-specific functions [62], are regulated by multiple upstream signalling pathways and as well as gene-specific roles in regulating estrogen-responsive changes in their protein levels or activity can effectively modulate genes in breast cancer cells [17,29,34]. It has also been gene expression. Unlike SRC-1 and SRC-3, which are overex- demonstrated that recruitment of a particular member of the pressed in different types of cancers, there are few reports SRC family to a DNA-associated transcription factor is deter- regarding a role of SRC-2 in oncogenesis [51,52]. In the present mined by the level of expression of that particular SRC, and PTMs study, we explored the potential function of SRC-2 in MCF-7 may alter the availability of SRCs [33]. breast cancer cells, and the role of PKA-mediated degradation of Activation of the PKA-signalling pathway specifically inhibits SRC-2 by characterization of the transcriptomes of SRC-2- the coactivator function of SRC-2 in MCF-7 cells by promoting its depleted MCF-7 cells and of cells treated with PKA-activating ubiquitination and degradation and thereby reducing ERa agents. We observed that downregulation of SRC-2 induces transactivation function [35,36]. In the present study we significant changes in the expression of several estrogen-responsive demonstrated that downregulation of SRC-2 by PKA modulates genes involved in breast cancer progression. Consistent with these the expression of several estrogen-responsive genes involved in findings, we observed that depletion of SRC-2 in MCF-7 cells breast tumourigenesis, including RET, BCAS1, EGR1, BCL11b and clearly stimulated proliferation of the cells. Together, the results TAGLN, indicating that PKA signalling through SRC-2 could be suggest an antiproliferative role of SRC-2 in MCF-7 cells. implicated in breast cancer progression. Moreover, functional A recent study also demonstrated that low levels of SRC-2 analyses of differentially expressed genes after cAMP-treatment expression in hepatocellular carcinoma patients were associated revealed enhanced expression of genes involved in cell cycle and with poor prognosis, and RNAi-mediated knockdown of NCOA2 in cell signalling processes, as well as genes encoding signalling diethylnitrosamine-treated mice promoted liver tumourigenesis molecules such as growth factors (data not shown). In line with [53]. Moreover, it has been reported that enhanced expression of this, we observed enhanced proliferation of MCF-7 cells treated SRC-2 in malignant pleural mesothelioma (MPM) tumour cells is with PKA activating agents. Adding cAMP elevating agents to the

PLOS ONE | www.plosone.org 7 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

Figure 4. Downregulation of SRC-2 promotes proliferation of MCF-7 cells. MCF-7 cells with KD of SRC-2 (SRC-2 shRNA) and control shRNA MCF-7 cells (Ctr shRNA) grown in phenol red-free DMEM supplemented with charcoaled stripped FBS (5%) for three days, were seeded in E-Plate 16 and treated with either Vehicle or 8-CPT-cAMP (150 mM), IBMX (50 mM) and forskolin (10 mM) (cAMP), in the absence or presence of 17b-estradiol (E2, 10 nM) (A and B, respectively). Proliferation of the cells was measured by using the xCELLigence system monitoring cellular impedance continuously for 120 hours. The results are representative of at least three independent experiments. doi:10.1371/journal.pone.0070096.g004

SRC-2-silenced cells resulted in the most pronounced proliferation Even though SRC-2 is classified as a coactivator we observed pattern. The gene expression and proliferation data suggest that that several genes also were up-regulated due to reduced SRC-2 PKA promotes tumourigenesis via both SRC-2-dependent- as well expression, suggesting a repressive effect of SRC-2 on the as -independent mechanisms. PKA is known to be implicated in expression of some estrogen-responsive genes in breast cancer initiation and progression of many tumours, but the mechanism by cells. However, the mechanism by which estrogen represses gene which PKA promote cancinogenesis is not clear [63]. The cAMP/ expression is unknown. Studies have shown that SRC-2 in contrast PKA signalling pathway appears to have both proliferative and to SRC-1 and SRC-3, possess a unique repression domain anti-proliferative effects on breast cancer cell growth [64–67], and encompassing GRIP1 amino acids 767–1006 utilized in repression various cAMP analogues have also been reported to exert different of GR-mediated inhibition of target gene expression [19]. It has effects on breast cancer cell growth which may be due to been suggested that the enhancing- or repressive effect of SRC-2 differences in their mechanisms of action [67]. on gene expression could be tissue specific and also to be dependent on the regulatory context of the target gene promoter.

PLOS ONE | www.plosone.org 8 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

Since we observed that depletion of SRC-2 led to enhanced Table S1 Primer sequences used for Q-rt-PCR. expression of several genes, including ERa-target genes, it is (DOC) possible that SRC-2 may also function as a transcriptional Table S2 Downregulated differentially expressed genes corepressor of ERa in breast cancer cells. induced by SRC-2 shRNA. FC $1.5, q-value = 0. Some studies have shown that the SRCs are able to compensate (XLSX) for the loss of individual SRC-member [68,69]. However, a recent ChIP-sequencing mapping study of the global chromatin binding Table S3 Upregulated differentially expressed genes sites of individual SRCs in MCF-7 cells, revealed limited degree of induced by SRC-2 shRNA. FC $1.5, q-value = 0. redundancy between the different SRCs [34], which has also been (XLSX) seen in a study comparing the role of the three SRC family Table S4 Downregulated differentially expressed genes members on metabolism in different organs [70]. By analysing the induced by cAMP. FC $1.5, q-value = 0. expression of selected SRC-2- and PKA-regulated (overlapping) (XLSX) genes in SRC-3-depleted MCF-7 cells, we confirmed a specific regulation of these genes by SRC-2 (data not shown). In addition, Table S5 Upregulated differentially expressed genes analyses of transcription factor binding sites in the promoter induced by cAMP. FC $1.5, q-value = 0. regions of the selected overlapping genes showed enriched NR2F (XLSX) binding motifs in their promoters, which have shown by others to Table S6 Overlapping downregulated differentially ex- be exclusively found for SRC-2 target genes [34]. Thus, we believe pressed genes induced by SRC-2 shRNA and cAMP. FC that the oncogenic phenotype observed in the SRC-2-depleted $1.3, q-value = 0. cells is SRC-2 specific and not caused by compensatory effects of (XLSX) the other SRCs. In summary, our data suggest that reduced levels of SRC-2 in Table S7 Upregulated overlapping differentially ex- breast cancer cells modulates the expression of estrogen-regulated pressed genes induced by SRC-2 shRNA and cAMP. FC genes leading to enhanced proliferation of breast cancer cells, $1.3, q-value = 0. suggesting that SRC-2 has antiproliferative properties in breast (XLSX) cancer cells. Abbreviations S1. (DOC) Supporting Information Figure S1 Correspondence analysis (CA) plot showing Acknowledgments projection of microarray samples. Ctr shRNA: pink Technical assistance from Carol Cook and Anita Ivarsflaten is highly squares, SRC-2 shRNA: green diamonds, cAMP: purple circles. appreciated. We thank the staff at the Norwegian Bioinformatics Platform The first principle component is shown on the x-axis and the and the Norwegian Microarray Consortium, University of Bergen Core second principle component is displayed on the y-axis. All three Facility for expert assistance with analyses. groups of samples are separated along the first principle component (22.706% component variance). The second principle Author Contributions component (15.192% component variance) separates the two Conceived and designed the experiments: ISF JVS GM. Performed the groups of treated samples (SRC-2 shRNA and cAMP) from the experiments: ISF TH. Analyzed the data: ISF TH MHF SND GM JVS. control group sample (Ctr shRNA). Contributed reagents/materials/analysis tools: ISF GM JVS. Wrote the (TIF) paper: ISF TH GM JVS.

References 1. Onate SA, Tsai SY, Tsai MJ, O’Malley BW (1995) Sequence and 10. Mukherjee A, Amato P, Craig-Allred D, DeMayo FJ, O’Malley BW, et al. (2007) characterization of a coactivator for the steroid hormone receptor superfamily. Steroid receptor coactivator 2: an essential coregulator of progestin-induced Science 270: 1354–1357. uterine and mammary morphogenesis. Ernst Schering Found Symp Proc: 55– 2. Hong H, Kohli K, Trivedi A, Johnson DL, Stallcup MR (1996) GRIP1, a novel 76. mouse protein that serves as a transcriptional coactivator in yeast for the 11. Jeong JW, Kwak I, Lee KY, White LD, Wang XP, et al. (2006) The genomic hormone binding domains of steroid receptors. Proc Natl Acad Sci U S A 93: analysis of the impact of steroid receptor coactivators ablation on hepatic 4948–4952. metabolism. Mol Endocrinol 20: 1138–1152. 3. Voegel JJ, Heine MJ, Zechel C, Chambon P, Gronemeyer H (1996) TIF2, a 160 12. Picard F, Gehin M, Annicotte J, Rocchi S, Champy MF, et al. (2002) SRC-1 kDa transcriptional mediator for the ligand-dependent activation function AF-2 and TIF2 control energy balance between white and brown adipose tissues. Cell of nuclear receptors. Embo J 15: 3667–3675. 111: 931–941. 4. Anzick SL, Kononen J, Walker RL, Azorsa DO, Tanner MM, et al. (1997) 13. Modder UI, Monroe DG, Fraser DG, Spelsberg TC, Rosen CJ, et al. (2009) AIB1, a steroid receptor coactivator amplified in breast and ovarian cancer. Skeletal consequences of deletion of steroid receptor coactivator-2/transcription Science 277: 965–968. intermediary factor-2. J Biol Chem 284: 18767–18777. 5. Chen H, Lin RJ, Schiltz RL, Chakravarti D, Nash A, et al. (1997) Nuclear 14. Reineke EL, York B, Stashi E, Chen X, Tsimelzon A, et al. (2012) SRC-2 receptor coactivator ACTR is a novel histone acetyltransferase and forms a Coactivator Deficiency Decreases Functional Reserve in Response to Pressure multimeric activation complex with P/CAF and CBP/p300. Cell 90: 569–580. Overload of Mouse Heart. PLoS One 7: e53395. 6. Li H, Gomes PJ, Chen JD (1997) RAC3, a steroid/nuclear receptor-associated 15. Jeong JW, Lee KY, Han SJ, Aronow BJ, Lydon JP, et al. (2007) The p160 coactivator that is related to SRC-1 and TIF2. Proc Natl Acad Sci U S A 94: steroid receptor coactivator 2, SRC-2, regulates murine endometrial function 8479–8484. and regulates progesterone-independent and -dependent gene expression. 7. Takeshita A, Cardona GR, Koibuchi N, Suen CS, Chin WW (1997) TRAM-1, Endocrinology 148: 4238–4250. A novel 160-kDa thyroid hormone receptor activator molecule, exhibits distinct 16. Li X, Wong J, Tsai SY, Tsai MJ, O’Malley BW (2003) Progesterone and properties from steroid receptor coactivator-1. J Biol Chem 272: 27629–27634. glucocorticoid receptors recruit distinct coactivator complexes and promote 8. Torchia J, Rose DW, Inostroza J, Kamei Y, Westin S, et al. (1997) The distinct patterns of local chromatin modification. Mol Cell Biol 23: 3763–3773. transcriptional co-activator p/CIP binds CBP and mediates nuclear-receptor 17. Won Jeong K, Chodankar R, Purcell DJ, Bittencourt D, Stallcup MR (2012) function. Nature 387: 677–684. Gene-specific patterns of coregulator requirements by estrogen receptor-alpha in 9. Gehin M, Mark M, Dennefeld C, Dierich A, Gronemeyer H, et al. (2002) The breast cancer cells. Mol Endocrinol 26: 955–966. function of TIF2/GRIP1 in mouse reproduction is distinct from those of SRC-1 18. Li S, Shang Y (2007) Regulation of SRC family coactivators by post- and p/CIP. Mol Cell Biol 22: 5923–5937. translational modifications. Cell Signal 19: 1101–1112.

PLOS ONE | www.plosone.org 9 July 2013 | Volume 8 | Issue 7 | e70096 SRC-2 and Estrogen-Responsive Genes in MCF-7 Cells

19. Rogatsky I, Luecke HF, Leitman DC, Yamamoto KR (2002) Alternate surfaces including TGFbeta1, PTEN, p53, and fibronectin. Cancer Gene Ther 13: 115– of transcriptional coregulator GRIP1 function in different glucocorticoid 124. receptor activation and repression contexts. Proc Natl Acad Sci U S A 99: 44. Kamimura K, Mishima Y, Obata M, Endo T, Aoyagi Y, et al. (2007) Lack of 16701–16706. Bcl11b tumor suppressor results in vulnerability to DNA replication stress and 20. Cvoro A, Tzagarakis-Foster C, Tatomer D, Paruthiyil S, Fox MS, et al. (2006) damages. Oncogene 26: 5840–5850. Distinct roles of unliganded and liganded estrogen receptors in transcriptional 45. Hino M, Doihara H, Kobayashi K, Aoe M, Shimizu N (2003) Caveolin-1 as repression. Mol Cell 21: 555–564. tumor suppressor gene in breast cancer. Surg Today 33: 486–490. 21. Rogatsky I, Zarember KA, Yamamoto KR (2001) Factor recruitment and 46. Collins C, Rommens JM, Kowbel D, Godfrey T, Tanner M, et al. (1998) TIF2/GRIP1 corepressor activity at a collagenase-3 response element that Positional cloning of ZNF217 and NABC1: genes amplified at 20q13.2 and mediates regulation by phorbol esters and hormones. EMBO J 20: 6071–6083. overexpressed in breast carcinoma. Proc Natl Acad Sci U S A 95: 8703–8708. 22. Uhlenhaut NH, Barish GD, Yu RT, Downes M, Karunasiri M, et al. (2013) 47. Eng C (1999) RET proto-oncogene in the development of human cancer. J Clin Insights into negative regulation by the glucocorticoid receptor from genome- Oncol 17: 380–393. wide profiling of inflammatory cistromes. Mol Cell 49: 158–171. 48. Morandi A, Plaza-Menacho I, Isacke CM (2011) RET in breast cancer: 23. Fleming FJ, Myers E, Kelly G, Crotty TB, McDermott EW, et al. (2004) functional and therapeutic implications. Trends Mol Med 17: 149–157. Expression of SRC-1, AIB1, and PEA3 in HER2 mediated endocrine resistant 49. Liang Z, Yoon Y, Votaw J, Goodman MM, Williams L, et al. (2005) Silencing of breast cancer; a predictive role for SRC-1. J Clin Pathol 57: 1069–1074. CXCR4 blocks breast cancer metastasis. Cancer Res 65: 967–971. 24. Murphy LC, Simon SL, Parkes A, Leygue E, Dotzlaw H, et al. (2000) Altered 50. Martinez A, Vos M, Guedez L, Kaur G, Chen Z, et al. (2002) The effects of expression of estrogen receptor coregulators during human breast tumorigenesis. adrenomedullin overexpression in breast tumor cells. J Natl Cancer Inst 94: Cancer Res 60: 6266–6271. 1226–1237. 25. Myers E, Fleming FJ, Crotty TB, Kelly G, McDermott EW, et al. (2004) Inverse 51. Walsh CA, Qin L, Tien JC, Young LS, Xu J (2012) The function of steroid relationship between ER-beta and SRC-1 predicts outcome in endocrine- receptor coactivator-1 in normal tissues and cancer. Int J Biol Sci 8: 470–485. resistant breast cancer. Br J Cancer 91: 1687–1693. 52. Ma G, Ren Y, Wang K, He J (2011) SRC-3 has a role in cancer other than as a 26. Qin L, Liao L, Redmond A, Young L, Yuan Y, et al. (2008) The AIB1 oncogene nuclear receptor coactivator. Int J Biol Sci 7: 664–672. promotes breast cancer metastasis by activation of PEA3-mediated matrix 53. O’Donnell KA, Keng VW, York B, Reineke EL, Seo D, et al. (2012) A Sleeping metalloproteinase 2 (MMP2) and MMP9 expression. Mol Cell Biol 28: 5937– Beauty mutagenesis screen reveals a tumor suppressor role for Ncoa2/Src-2 in 5950. liver cancer. Proc Natl Acad Sci U S A 109: E1377–1386. 27. Qin L, Liu Z, Chen H, Xu J (2009) The steroid receptor coactivator-1 regulates 54. Jennings CJ, O’Grady A, Cummins R, Murer B, Al-Alawi M, et al. (2012) twist expression and promotes breast cancer metastasis. Cancer Res 69: 3819– Sustained expression of steroid receptor coactivator SRC-2/TIF-2 is associated 3827. with better prognosis in malignant pleural mesothelioma. J Thorac Oncol 7: 28. Zhao C, Yasui K, Lee CJ, Kurioka H, Hosokawa Y, et al. (2003) Elevated 243–248. expression levels of NCOA3, TOP1, and TFAP2C in breast tumors as 55. Grivas PD, Tzelepi V, Sotiropoulou-Bonikou G, Kefalopoulou Z, Papavassiliou predictors of poor prognosis. Cancer 98: 18–23. AG, et al. (2009) Estrogen receptor alpha/beta, AIB1, and TIF2 in colorectal 29. Karmakar S, Foster EA, Smith CL (2009) Unique roles of p160 coactivators for carcinogenesis: do coregulators have prognostic significance? Int J Colorectal Dis regulation of breast cancer cell proliferation and estrogen receptor-alpha 24: 613–622. transcriptional activity. Endocrinology 150: 1588–1596. 56. Gregory CW, He B, Johnson RT, Ford OH, Mohler JL, et al. (2001) A mechanism for androgen receptor-mediated prostate cancer recurrence after 30. Planas-Silva MD, Shang Y, Donaher JL, Brown M, Weinberg RA (2001) AIB1 androgen deprivation therapy. Cancer Res 61: 4315–4319. enhances estrogen-dependent induction of cyclin D1 expression. Cancer Res 61: 57. Kershah SM, Desouki MM, Koterba KL, Rowan BG (2004) Expression of 3858–3862. estrogen receptor coregulators in normal and malignant human endometrium. 31. List HJ, Lauritsen KJ, Reiter R, Powers C, Wellstein A, et al. (2001) Ribozyme Gynecol Oncol 92: 304–313. targeting demonstrates that the nuclear receptor coactivator AIB1 is a rate- 58. Kefalopoulou Z, Tzelepi V, Zolota V, Grivas PD, Christopoulos C, et al. (2012) limiting factor for estrogen-dependent growth of human MCF-7 breast cancer Prognostic value of novel biomarkers in astrocytic brain tumors: nuclear receptor cells. J Biol Chem 276: 23763–23768. co-regulators AIB1, TIF2, and PELP1 are associated with high tumor grade and 32. Xu J, Wu RC, O’Malley BW (2009) Normal and cancer-related functions of the worse patient prognosis. J Neurooncol 106: 23–31. p160 steroid receptor co-activator (SRC) family. Nat Rev Cancer 9: 615–630. 59. Haugan Moi LL, Hauglid Flageng M, Gandini S, Guerrieri-Gonzaga A, 33. Zhang H, Yi X, Sun X, Yin N, Shi B, et al. (2004) Differential gene regulation Bonanni B, et al. (2010) Effect of low-dose tamoxifen on steroid receptor by the SRC family of coactivators. Genes Dev 18: 1753–1765. coactivator 3/amplified in breast cancer 1 in normal and malignant human 34. Zwart W, Theodorou V, Kok M, Canisius S, Linn S, et al. (2011) Oestrogen breast tissue. Clin Cancer Res 16: 2176–2186. receptor-co-factor-chromatin specificity in the transcriptional regulation of 60. Flageng MH, Moi LL, Dixon JM, Geisler J, Lien EA, et al. (2009) Nuclear breast cancer. EMBO J 30: 4764–4776. receptor co-activators and HER-2/neu are upregulated in breast cancer patients 35. Fenne IS, Hoang T, Hauglid M, Sagen JV, Lien EA, et al. (2008) Recruitment of during neo-adjuvant treatment with aromatase inhibitors. Br J Cancer 101: coactivator glucocorticoid receptor interacting protein 1 to an estrogen receptor 1253–1260. transcription complex is regulated by the 39,59-cyclic adenosine 5’-monophos- 61. Zou JX, Zhong Z, Shi XB, Tepper CG, deVere White RW, et al. (2006) phate-dependent protein kinase. Endocrinology 149: 4336–4345. ACTR/AIB1/SRC-3 and androgen receptor control prostate cancer cell 36. Hoang T, Fenne IS, Cook C, Borud B, Bakke M, et al. (2004) cAMP-dependent proliferation and tumor growth through direct control of cell cycle genes. protein kinase regulates ubiquitin-proteasome-mediated degradation and Prostate 66: 1474–1486. subcellular localization of the nuclear receptor coactivator GRIP1. J Biol Chem 62. Han SJ, DeMayo FJ, Xu J, Tsai SY, Tsai MJ, et al. (2006) Steroid receptor 279: 49120–49130. coactivator (SRC)-1 and SRC-3 differentially modulate tissue-specific activation 37. Hoang T, Fenne IS, Madsen A, Bozickovic O, Johannessen M, et al. (2013) functions of the progesterone receptor. Mol Endocrinol 20: 45–55. cAMP Response Element-Binding Protein Interacts With and Stimulates the 63. Mucignat-Caretta ACaC (2011) Protein Kinase A in Cancer. Cancers. 913–926. Proteasomal Degradation of the Nuclear Receptor Coactivator GRIP1. 64. Al-Dhaheri MH, Rowan BG (2006) PKA exhibits selective modulation of Endocrinology 154: 1513–1527. estradiol dependent transcription in breast cancer cells that is associated with 38. Bolstad BM, Irizarry RA, Astrand M, Speed TP (2003) A comparison of decreased ligand binding, altered ER{alpha} promoter interaction and changes normalization methods for high density oligonucleotide array data based on in receptor phosphorylation. Mol Endocrinol. variance and bias. Bioinformatics 19: 185–193. 65. Dudek P, Picard D (2008) Genomics of signaling crosstalk of estrogen receptor 39. Fellenberg K, Hauser NC, Brors B, Neutzner A, Hoheisel JD, et al. (2001) alpha in breast cancer cells. PLoS One 3: e1859. Correspondence analysis applied to microarray data. Proc Natl Acad Sci U S A 66. Kung W, Roos W, Eppenberger U (1983) Growth stimulation of human breast 98: 10781–10786. cancer MCF-7 cells by dibutyryl cyclic AMP. Cell Biol Int Rep 7: 345–351. 40. Tusher VG, Tibshirani R, Chu G (2001) Significance analysis of microarrays 67. Lamb D, Steinberg RA (2002) Anti-proliferative effects of 8-chloro-cAMP and applied to the ionizing radiation response. Proc Natl Acad Sci U S A 98: 5116– other cAMP analogs are unrelated to their effects on protein kinase A regulatory 5121. subunit expression. J Cell Physiol 192: 216–224. 41. Borud B, Hoang T, Bakke M, Jacob AL, Lund J, et al. (2002) The nuclear 68. Xu J, Li Q (2003) Review of the in vivo functions of the p160 steroid receptor receptor coactivators p300/CBP/cointegrator-associated protein (p/CIP) and coactivator family. Mol Endocrinol 17: 1681–1692. transcription intermediary factor 2 (TIF2) differentially regulate PKA-stimulated 69. Tien JC, Zhou S, Xu J (2009) The role of SRC-1 in murine prostate transcriptional activity of steroidogenic factor 1. Mol Endocrinol 16: 757–773. cancinogenesis is nonessential due to a possible compensation of SRC-3/AIB1 42. Assinder SJ, Stanton JA, Prasad PD (2009) Transgelin: an actin-binding protein overexpression. Int J Biol Sci 5: 256–264. and tumour suppressor. Int J Biochem Cell Biol 41: 482–486. 70. York B, Sagen JV, Tsimelzon A, Louet JF, Chopra AR, et al. (2013) Research 43. Baron V, Adamson ED, Calogero A, Ragona G, Mercola D (2006) The resource: tissue- and pathway-specific metabolomic profiles of the steroid transcription factor Egr1 is a direct regulator of multiple tumor suppressors receptor coactivator (SRC) family. Mol Endocrinol 27: 366–380.

PLOS ONE | www.plosone.org 10 July 2013 | Volume 8 | Issue 7 | e70096